Digital multimedia signal processing via AIoT and its application in smart home
Wang Peng · Internet Technology Letters · 2021
Abstract The popularity of the Internet of Things (IoT) makes a large number of vulnerable devices connected to the Internet, which brings a large number of digital multimedia signals. How to process these digital multimedia signals and ensure the security in IoT has become an important topic in the IoT applications and services. With the development of data analysis and artificial intelligence, the IoT devices present intelligent behaviors and become the new generation of information infrastructure. This paper proposed a framework to process the digital signals generated by IoT devices and ensure the IoT security. First, the original traffic data passes through the security gateway and the associated traffic data is collected by a damping window. Second, the collected traffic data is processed by a feature analysis module to construct a fingerprint library. Third, the fingerprint library is used to learn a machine learning model. Lastly, the learnt machine learning model is used to determine whether the IoT network exists abnormal behaviors. The proposed framework is evaluated on ISOT Botnet dataset.